1. Field of the Invention
The present invention relates to a system, a method and a program for structured document derivation for converting a structured document from one representation to another without changing the contents of the document.
2. Description of the Prior Art
An example of a conventional structured document test generating system is disclosed in Japanese Patent Application laid open No. 2003-280942. FIG. 1 is a block diagram showing the conventional structured document test generating system. As shown in FIG. 1, the conventional structured document test generating system comprises a structured document type defining document inputting section 11, a structured document type defining document analyzer 12 and a test program generator 13. Such conventional structured document test generating system operates as follows. A structured document type defining document 10 such as DTD (Document Type Definition) is input to the structured document type defining document inputting section 11 to be analyzed in the structured document type defining document analyzer 12. Based on the analysis result, the test program generator 13 generates a test program 15 of the structured document.
Further, Japanese Patent Application laid open No. 2002-342316 discloses an example of a conventional structured document conversion system. FIG. 2 is a block diagram showing the conventional structured document conversion system. As shown in FIG. 2, the conventional structured document conversion system comprises a structured document type conversion defining document inputting section 22, a structured document inputting section 23 and a structured document converter 24. Such conventional structured document conversion system operates as follows. A structured document type conversion defining document 21 is input to the structured document type conversion defining document inputting section 22. According to the rule of the input document, the structured document converter 24 converts the structure of a structured document 20 input from the structured document inputting section 23 to output a purposive form structured document 25 different from the input structured document 20.
The first problem is that the structured document conversion system has not generated structured documents of different representations having the same contents. With a structured document, because it is possible to have the same data in common among a plurality of systems, systems for document structure conversion have been highly developed for accepting the difference among document structures in the systems. On the other hand, it has not been conceived to convert a structured document to derivatives in different representations having the same contents. Incidentally, “structured documents of different representations having the same contents” indicate structured documents the same in structure and information therein and different in output representation such as name of namespace, space for indentation, or character code.
The second problem is that the structured document test generating system has focused on generating, by using a structured document type defining document, a structured document meeting the definition to accept it, and has not allowed enough for the variation of representations of a structured document. In a conventional test generating system, while the conversion of contents and structure of input data is an important test item, it has never been thought about whether or not structured documents generally showing a diversity in representation of the same contents of data are all acceptable.
The third problem is that structured document processing systems have not worked together for automatically avoiding structured documents having a vulnerable pattern by inputting test results indicating the diversity of representation of a structured document to a system for automatically compensating weakness in the performance of the systems. This is because there has been neither a system focusing on the diversity of representation of a structured document nor consideration for the use of the test results as to the diversity of representation for automatically compensating weakness in the performance of the systems in runtime.
The fourth problem is that a system has not invented for improving the quality of processes of a structured document processing system by providing an intermediate node. The intermediate node has been considered to mainly perform logging, routing and representation conversion and the use of the intermediate node has not been considered to improve the end service without changing data contents.